[ { "loss": 0.2436, "grad_norm": 0.23789972066879272, "learning_rate": 6.666666666666667e-05, "epoch": 0.011494252873563218, "step": 10 }, { "loss": 0.0853, "grad_norm": 0.12851466238498688, "learning_rate": 0.00014074074074074076, "epoch": 0.022988505747126436, "step": 20 }, { "loss": 0.0627, "grad_norm": 0.16485165059566498, "learning_rate": 0.0001999972223815946, "epoch": 0.034482758620689655, "step": 30 }, { "loss": 0.048, "grad_norm": 0.08059470355510712, "learning_rate": 0.00019990002193828922, "epoch": 0.04597701149425287, "step": 40 }, { "loss": 0.0421, "grad_norm": 0.10289835184812546, "learning_rate": 0.0001996640948394666, "epoch": 0.05747126436781609, "step": 50 }, { "loss": 0.05, "grad_norm": 0.10482244938611984, "learning_rate": 0.00019928976870644988, "epoch": 0.06896551724137931, "step": 60 }, { "loss": 0.0377, "grad_norm": 0.10887732356786728, "learning_rate": 0.00019877756334905984, "epoch": 0.08045977011494253, "step": 70 }, { "loss": 0.0427, "grad_norm": 0.10280067473649979, "learning_rate": 0.00019812819004377862, "epoch": 0.09195402298850575, "step": 80 }, { "loss": 0.0375, "grad_norm": 0.1345912218093872, "learning_rate": 0.0001973425505460321, "epoch": 0.10344827586206896, "step": 90 }, { "loss": 0.0357, "grad_norm": 0.3418385982513428, "learning_rate": 0.00019642173583796267, "epoch": 0.11494252873563218, "step": 100 }, { "eval_loss": 0.052477214485406876, "eval_runtime": 20.6792, "eval_samples_per_second": 14.507, "eval_steps_per_second": 1.838, "epoch": 0.11494252873563218, "step": 100 }, { "loss": 0.047, "grad_norm": 0.07799140363931656, "learning_rate": 0.00019536702461343168, "epoch": 0.12643678160919541, "step": 110 }, { "loss": 0.0327, "grad_norm": 0.1051594689488411, "learning_rate": 0.00019417988150235494, "epoch": 0.13793103448275862, "step": 120 }, { "loss": 0.0418, "grad_norm": 0.08026126027107239, "learning_rate": 0.00019286195503683708, "epoch": 0.14942528735632185, "step": 130 }, { "loss": 0.037, "grad_norm": 0.07206742465496063, "learning_rate": 0.0001914150753619294, "epoch": 0.16091954022988506, "step": 140 }, { "loss": 0.033, "grad_norm": 0.1900654435157776, "learning_rate": 0.00018984125169419007, "epoch": 0.1724137931034483, "step": 150 }, { "loss": 0.0348, "grad_norm": 0.11883541941642761, "learning_rate": 0.00018814266953157556, "epoch": 0.1839080459770115, "step": 160 }, { "loss": 0.0296, "grad_norm": 0.11700455099344254, "learning_rate": 0.00018632168761853826, "epoch": 0.19540229885057472, "step": 170 }, { "loss": 0.0305, "grad_norm": 0.15726420283317566, "learning_rate": 0.00018438083467054443, "epoch": 0.20689655172413793, "step": 180 }, { "loss": 0.0319, "grad_norm": 0.18836188316345215, "learning_rate": 0.00018232280586256102, "epoch": 0.21839080459770116, "step": 190 }, { "loss": 0.0259, "grad_norm": 0.08308509737253189, "learning_rate": 0.0001801504590863877, "epoch": 0.22988505747126436, "step": 200 }, { "eval_loss": 0.02992151863873005, "eval_runtime": 20.3901, "eval_samples_per_second": 14.713, "eval_steps_per_second": 1.864, "epoch": 0.22988505747126436, "step": 200 }, { "loss": 0.0313, "grad_norm": 0.05260265991091728, "learning_rate": 0.00017786681098203158, "epoch": 0.2413793103448276, "step": 210 }, { "loss": 0.0229, "grad_norm": 0.0633602887392044, "learning_rate": 0.00017547503274863496, "epoch": 0.25287356321839083, "step": 220 }, { "loss": 0.028, "grad_norm": 0.11880059540271759, "learning_rate": 0.00017297844574077446, "epoch": 0.26436781609195403, "step": 230 }, { "loss": 0.0264, "grad_norm": 0.05497035011649132, "learning_rate": 0.0001703805168562453, "epoch": 0.27586206896551724, "step": 240 }, { "loss": 0.0213, "grad_norm": 0.07999792695045471, "learning_rate": 0.00016768485372173697, "epoch": 0.28735632183908044, "step": 250 }, { "loss": 0.0284, "grad_norm": 0.0921878069639206, "learning_rate": 0.00016489519968308457, "epoch": 0.2988505747126437, "step": 260 }, { "loss": 0.0216, "grad_norm": 0.057495057582855225, "learning_rate": 0.00016201542860705336, "epoch": 0.3103448275862069, "step": 270 }, { "loss": 0.0235, "grad_norm": 0.0719340518116951, "learning_rate": 0.00015904953950187457, "epoch": 0.3218390804597701, "step": 280 }, { "loss": 0.0232, "grad_norm": 0.11424289643764496, "learning_rate": 0.00015600165096400307, "epoch": 0.3333333333333333, "step": 290 }, { "loss": 0.0185, "grad_norm": 0.11561417579650879, "learning_rate": 0.0001528759954588082, "epoch": 0.3448275862068966, "step": 300 }, { "eval_loss": 0.034754782915115356, "eval_runtime": 20.413, "eval_samples_per_second": 14.697, "eval_steps_per_second": 1.862, "epoch": 0.3448275862068966, "step": 300 }, { "loss": 0.029, "grad_norm": 0.08098062872886658, "learning_rate": 0.00014967691344313996, "epoch": 0.3563218390804598, "step": 310 }, { "loss": 0.021, "grad_norm": 0.07463961839675903, "learning_rate": 0.00014640884733793216, "epoch": 0.367816091954023, "step": 320 }, { "loss": 0.0247, "grad_norm": 0.055279821157455444, "learning_rate": 0.00014307633535921317, "epoch": 0.3793103448275862, "step": 330 }, { "loss": 0.0203, "grad_norm": 0.18279875814914703, "learning_rate": 0.00013968400521608969, "epoch": 0.39080459770114945, "step": 340 }, { "loss": 0.0151, "grad_norm": 0.07882044464349747, "learning_rate": 0.0001362365676844557, "epoch": 0.40229885057471265, "step": 350 }, { "loss": 0.0261, "grad_norm": 0.07913503050804138, "learning_rate": 0.00013273881006535074, "epoch": 0.41379310344827586, "step": 360 }, { "loss": 0.0176, "grad_norm": 0.06874211877584457, "learning_rate": 0.00012919558953705055, "epoch": 0.42528735632183906, "step": 370 }, { "loss": 0.0187, "grad_norm": 0.06528666615486145, "learning_rate": 0.0001256118264101228, "epoch": 0.4367816091954023, "step": 380 }, { "loss": 0.0158, "grad_norm": 0.08248886466026306, "learning_rate": 0.00012199249729481357, "epoch": 0.4482758620689655, "step": 390 }, { "loss": 0.0132, "grad_norm": 0.24866768717765808, "learning_rate": 0.00011834262819025326, "epoch": 0.45977011494252873, "step": 400 }, { "eval_loss": 0.018751109018921852, "eval_runtime": 20.2674, "eval_samples_per_second": 14.802, "eval_steps_per_second": 1.875, "epoch": 0.45977011494252873, "step": 400 }, { "loss": 0.0212, "grad_norm": 0.11699023842811584, "learning_rate": 0.00011466728750507805, "epoch": 0.47126436781609193, "step": 410 }, { "loss": 0.0176, "grad_norm": 0.08753076195716858, "learning_rate": 0.00011097157901915953, "epoch": 0.4827586206896552, "step": 420 }, { "loss": 0.0226, "grad_norm": 0.10365070402622223, "learning_rate": 0.00010726063479621573, "epoch": 0.4942528735632184, "step": 430 }, { "loss": 0.0189, "grad_norm": 0.18819937109947205, "learning_rate": 0.00010353960805714566, "epoch": 0.5057471264367817, "step": 440 }, { "loss": 0.0129, "grad_norm": 0.10421279072761536, "learning_rate": 9.98136660239837e-05, "epoch": 0.5172413793103449, "step": 450 }, { "loss": 0.0224, "grad_norm": 0.07066013664007187, "learning_rate": 9.60879827444116e-05, "epoch": 0.5287356321839081, "step": 460 }, { "loss": 0.0158, "grad_norm": 0.04448763281106949, "learning_rate": 9.236773190679163e-05, "epoch": 0.5402298850574713, "step": 470 }, { "loss": 0.0216, "grad_norm": 0.06225728988647461, "learning_rate": 8.865807965569879e-05, "epoch": 0.5517241379310345, "step": 480 }, { "loss": 0.022, "grad_norm": 0.12085612863302231, "learning_rate": 8.496417741792913e-05, "epoch": 0.5632183908045977, "step": 490 }, { "loss": 0.0152, "grad_norm": 0.04715719074010849, "learning_rate": 8.129115474894523e-05, "epoch": 0.5747126436781609, "step": 500 }, { "eval_loss": 0.01953456737101078, "eval_runtime": 20.3994, "eval_samples_per_second": 14.706, "eval_steps_per_second": 1.863, "epoch": 0.5747126436781609, "step": 500 }, { "loss": 0.0214, "grad_norm": 0.08593538403511047, "learning_rate": 7.764411220969414e-05, "epoch": 0.5862068965517241, "step": 510 }, { "loss": 0.0174, "grad_norm": 0.08397146314382553, "learning_rate": 7.402811428368832e-05, "epoch": 0.5977011494252874, "step": 520 }, { "loss": 0.0163, "grad_norm": 0.07681377232074738, "learning_rate": 7.044818234418576e-05, "epoch": 0.6091954022988506, "step": 530 }, { "loss": 0.0135, "grad_norm": 0.061246950179338455, "learning_rate": 6.69092876812353e-05, "epoch": 0.6206896551724138, "step": 540 }, { "loss": 0.0111, "grad_norm": 0.1086721196770668, "learning_rate": 6.341634459827052e-05, "epoch": 0.632183908045977, "step": 550 }, { "loss": 0.0185, "grad_norm": 0.10877221822738647, "learning_rate": 5.9974203587837986e-05, "epoch": 0.6436781609195402, "step": 560 }, { "loss": 0.0134, "grad_norm": 0.25607386231422424, "learning_rate": 5.658764459593691e-05, "epoch": 0.6551724137931034, "step": 570 }, { "loss": 0.0155, "grad_norm": 0.06046198308467865, "learning_rate": 5.3261370384323984e-05, "epoch": 0.6666666666666666, "step": 580 }, { "loss": 0.0156, "grad_norm": 0.08047638833522797, "learning_rate": 5.000000000000002e-05, "epoch": 0.6781609195402298, "step": 590 }, { "loss": 0.0118, "grad_norm": 0.09806278347969055, "learning_rate": 4.680806236094756e-05, "epoch": 0.6896551724137931, "step": 600 }, { "eval_loss": 0.016160501167178154, "eval_runtime": 20.2233, "eval_samples_per_second": 14.834, "eval_steps_per_second": 1.879, "epoch": 0.6896551724137931, "step": 600 }, { "loss": 0.0154, "grad_norm": 0.12618760764598846, "learning_rate": 4.3689989967026935e-05, "epoch": 0.7011494252873564, "step": 610 }, { "loss": 0.0105, "grad_norm": 0.16422143578529358, "learning_rate": 4.065011274476349e-05, "epoch": 0.7126436781609196, "step": 620 }, { "loss": 0.0155, "grad_norm": 0.3045821487903595, "learning_rate": 3.769265203457375e-05, "epoch": 0.7241379310344828, "step": 630 }, { "loss": 0.0158, "grad_norm": 0.12457603216171265, "learning_rate": 3.482171472878062e-05, "epoch": 0.735632183908046, "step": 640 }, { "loss": 0.0096, "grad_norm": 0.04488776624202728, "learning_rate": 3.20412875685569e-05, "epoch": 0.7471264367816092, "step": 650 }, { "loss": 0.0164, "grad_norm": 0.05114804208278656, "learning_rate": 2.935523160771786e-05, "epoch": 0.7586206896551724, "step": 660 }, { "loss": 0.0115, "grad_norm": 0.03534635156393051, "learning_rate": 2.6767276851049816e-05, "epoch": 0.7701149425287356, "step": 670 }, { "loss": 0.0138, "grad_norm": 0.06601171940565109, "learning_rate": 2.4281017074620584e-05, "epoch": 0.7816091954022989, "step": 680 }, { "loss": 0.0142, "grad_norm": 0.10086585581302643, "learning_rate": 2.1899904835264896e-05, "epoch": 0.7931034482758621, "step": 690 }, { "loss": 0.0098, "grad_norm": 0.11490526050329208, "learning_rate": 1.962724667617436e-05, "epoch": 0.8045977011494253, "step": 700 }, { "eval_loss": 0.012033017352223396, "eval_runtime": 20.3491, "eval_samples_per_second": 14.743, "eval_steps_per_second": 1.867, "epoch": 0.8045977011494253, "step": 700 }, { "loss": 0.0183, "grad_norm": 0.17799951136112213, "learning_rate": 1.746619853524991e-05, "epoch": 0.8160919540229885, "step": 710 }, { "loss": 0.0125, "grad_norm": 0.1625465303659439, "learning_rate": 1.5419761362593332e-05, "epoch": 0.8275862068965517, "step": 720 }, { "loss": 0.0111, "grad_norm": 0.04061046242713928, "learning_rate": 1.3490776953223105e-05, "epoch": 0.8390804597701149, "step": 730 }, { "loss": 0.0102, "grad_norm": 0.026591680943965912, "learning_rate": 1.1681924000801791e-05, "epoch": 0.8505747126436781, "step": 740 }, { "loss": 0.0098, "grad_norm": 0.04456641525030136, "learning_rate": 9.995714377854936e-06, "epoch": 0.8620689655172413, "step": 750 }, { "loss": 0.0162, "grad_norm": 0.19221937656402588, "learning_rate": 8.434489647647092e-06, "epoch": 0.8735632183908046, "step": 760 }, { "loss": 0.0099, "grad_norm": 0.04924648255109787, "learning_rate": 7.000417812558302e-06, "epoch": 0.8850574712643678, "step": 770 }, { "loss": 0.0127, "grad_norm": 0.09919004142284393, "learning_rate": 5.695490303477124e-06, "epoch": 0.896551724137931, "step": 780 }, { "loss": 0.0111, "grad_norm": 0.04783658683300018, "learning_rate": 4.521519214390257e-06, "epoch": 0.9080459770114943, "step": 790 }, { "loss": 0.0094, "grad_norm": 0.11320256441831589, "learning_rate": 3.4801347860093924e-06, "epoch": 0.9195402298850575, "step": 800 }, { "eval_loss": 0.013403651304543018, "eval_runtime": 20.315, "eval_samples_per_second": 14.767, "eval_steps_per_second": 1.871, "epoch": 0.9195402298850575, "step": 800 }, { "loss": 0.016, "grad_norm": 0.32974982261657715, "learning_rate": 2.572783141929469e-06, "epoch": 0.9310344827586207, "step": 810 }, { "loss": 0.0095, "grad_norm": 0.24235671758651733, "learning_rate": 1.8007242804619628e-06, "epoch": 0.9425287356321839, "step": 820 }, { "loss": 0.01, "grad_norm": 0.08694616705179214, "learning_rate": 1.165030324932137e-06, "epoch": 0.9540229885057471, "step": 830 }, { "loss": 0.0113, "grad_norm": 0.09477345645427704, "learning_rate": 6.665840348696994e-07, "epoch": 0.9655172413793104, "step": 840 }, { "loss": 0.0086, "grad_norm": 0.048642598092556, "learning_rate": 3.060775801604354e-07, "epoch": 0.9770114942528736, "step": 850 }, { "loss": 0.0148, "grad_norm": 0.06935633718967438, "learning_rate": 8.401157986109853e-08, "epoch": 0.9885057471264368, "step": 860 }, { "loss": 0.0087, "grad_norm": 0.06433246284723282, "learning_rate": 6.944070123582336e-10, "epoch": 1.0, "step": 870 }, { "train_runtime": 6469.3899, "train_samples_per_second": 4.303, "train_steps_per_second": 0.134, "total_flos": 7.495713730424996e+17, "train_loss": 0.024751038421159503, "epoch": 1.0, "step": 870 } ]